PSI - Issue 2_A

Daiki Shiozawa et al. / Procedia Structural Integrity 2 (2016) 2091–2096 Author name / Structural Integrity Procedia 00 (2016) 000–000

2093

3

D (2) In the fatigue limit estimation based on dissipated energy, the dissipated energy is measured for various levels of stress amplitude during short time cyclic loading (staircase-like stress level test). It is known that the change of dissipated energy shows sharp increase from a certain stress level, as shown in Fig. 2. The estimation scheme is summarized as follows (Irie and Inoue, (2010)); (1) The measured results are divided into two groups, and an approximate straight line is generated for each group. The residual sum of squares for each group is calculated about some grouping. The grouping which minimizes the residual sum of squares is decided as the suitable one. (2) The point where the two approximate straight lines cross is determined as the estimated fatigue limit. 3. Experimental set-up The material under test is JIS type 316L austenitic stainless steel. The geometry of specimen is shown in Fig. 3. The specimens were subjected to unidirectional pre-straining at levels of 6% and 15% (total strain). Full reversed cyclic axis loading ( R =-1) with a frequency of 5 Hz was applied to the specimen by the electrohydraulic servo fatigue testing machine. Temperature change on the specimen surface was measured by infrared thermography with a MCT array detector. Dissipated energy are obtained from measured temperature change by two types of lock-in algorism, one using basically Fourier analysis and the other using Fourier analysis with phase information of dissipated energy (phase 2 f lock-in infrared method) (3) . The double frequency component of temperature change on the specimen includes the influence of the energy dissipation and harmonic vibration of the fatigue testing machine. The phase difference between the temperature change due to the energy dissipation and the wave with double frequency of thermoelastic temperature change  is a specific value (Shiozawa and Sakagami, (2015)). The phase 2 f lock-in infrared method utilizes a specific phase of the dissipated energy, as following equation; ρ q c T   where f load and f meas are the load frequency and the measurement frequency of the thermal camera, respectively. A correlation with a negative value is set to zero. The phase 2 f lock-in infrared method is effective for removing the noise component such as the thermoelastic temperature change due to the harmonic vibration of fatigue testing machine. 4. Results and discussion S - N curves for reference specimen (0% specimen) and pre-strained specimen (6% specimen) are shown in Fig. 4. The fatigue limit for 0% and 6% specimens are determined around 250 and 315MPa from Fig. 4, respectively. The     load D 0 meas 2 ˆ ˆ sin 2 2      N E t f T   T t t   N f                (3)

Fig. 3 Geometry of specimen.

Fig. 4 S - N curves